A new dynamic simulation model for buildings has been developed which emerges by itself as it receivesdata or signals from the building under consideration. The model exists in two forms: Software written for the PC Windows environment, and Hardware in the form of a silicon microchip. This paper describes the background of the mathematical model and compares the results of its simulations with those of a conventional simulation model, in both cases using data from monitoring of an existing building. New possibilities for the development of advanced control systems are discussed.

This paper describes an artificial neural network- (ANN) modeling approach forpredicting the energv anddemand savings resulting from energy conservation measure (ECM) retrofits in select buildings. Simulated data sequences were used to minimize the experimental uncertainty in the initial model, and to provide post period data for training. A university building was chosen to provide the data set for this study. The building was modeled and calibrated using the DOE-2.IE building energy analysis program.

There is increasing interest in Europe in the use of displacement ventilation and chilled ceiling coolingsystems. A modelling methodology is presented here that deals with the significantly different heat transfer characteristics of these systems compared with conventional all-air systems. The purpose of the work is to develop a room model that is computationally efficient enough for annual hourly simulation purposes and a nodal model has been developed that is intermediate in complexity between a single air node model and a CFD model.

This paper presents the concept of an "open" simulation environment for performance-driven designexploration as a multi-directional approach to computer-aided daylighting modeling. A prototypicalrealization of a Generative Simulation Tool for Architectural Lighting (GESTALT) for simultaneoustreatment of daylighting-related design and performance variables is introduced. Earlier studies demonstrated that GESTALT can operate in an "explicit" mode, using a fast-response computational module.

Dealing with issues of acquiring and accessing design knowledge in the conceptual stage of the design process is the focus of this research. This research starts by presenting a brief background about the limitations of the available energy-based CAAD tools. It then provides an illustration and description of the entire architecture of the conceptual model, identifies its different components and explains the relationships and interactions among these components.

This paper deals with the problem of empirical validation of thermal performance computer programs. It begins with a brief review of a number of techniques which have been used as a measure of the goodness-of-fit between measured and predicted data in a variety of empirical validation exercises. Several inadequacies inherent in existing techniques are identified as, a) no attempt is made to take into account the severity of the validation test. b) none give a single measure of the success (or otherwise) of the test. c) isolation of sources of error are difficult.

From a structural perspective, the question is whether temperature changes during the lifetime of a building are sufficient to affect its integrity. Changes in average temperature and temperature differentials are the key thermal loading parameters. Ile purpose of this paper is to show bow computer models have been used to construct design charts for thermal loading. Firstly, the models were calibrated from temperature measurements of actual structures. The field tests cover several years and a range of climatic conditions.

After a brief review of the development of thermal design tools world-wide, those available in Australia are considered and four are selected for validation. A review of validation methodologies is followed by simulation results compared with measured data obtained from simple test cells. Some preliminary results are presented The current continuation of the work is outlined.

A major case study of a high rise, city-centre of ce building in continental Europe was undertaken in a true commercial environment. Three dimensional thermal and Computational Fluid Dynamics (CFD) modelling was carried out for major spaces in parallel with wind tunnel tests and results compared. Three different methods were used to assess the ventilation rate in the wind tunnel tests: surface pressure coefficients, tracer gas and direct velocity measurements.

Me design of a fuzzy rule-based controller for the mixing-box of an air-handling unit is used to demonstrate how data obtained by computer simulation can be used to generate the rules. The controller uses measurements of the return and fresh air temperatures, together with the pressure drops across the dampers, to calculate values for the actuator control signals which determine the position of the dampers. The paper explains the methods used to generate the rules, discusses the choice of fuzzy reference sets and describes the approach used to produce suitable training data.